XVIIT.—Researches in Thermometry. By Epmunp J. Mitts, D.Sc., F.R.S. 
Communicated by Professor Sir WiLL1AM THomson, D.C.L., F.R.S. 
(Read May 3, 1880.) 
In the course of a series of experiments upon fusion point, as determined 
by the mercurial thermometer, I had occasion to investigate several of the 
properties of that instrument. In the hope that the results of that investiga- 
tion may economise the labour of other experimenters, and aid in its accuracy, 
I have ventured to submit them to the consideration of the Society. The 
memoir comprises the following sections :— 
I. Calibration and the calibration unit. 
IJ. The exposure correction. 
III. The movement of the zero with (a) time, and (8) temperature. 
IV. PoGcenporrr’s correction. 
V. Comparison of the mercurial with the air thermometer. 
VI. Effect of compression. 
I, CALIBRATION AND THE CALIBRATION UNIT. 
Various methods have been proposed for calibrating finished thermometers, 
all agreeing in using a small thread of mercury, say twenty scale divisions long, 
or less, as a calibration unit. I find it a much easier plan to calibrate the 
graduated stem before the bulb is blown on, For this purpose the stem is 
placed in a horizontal position, parallel to a horizontal kathetometer, By 
means of a piece of india-rubber tubing attached to one of its ends, a thread of 
mercury of suitable length can be drawn in at the other end; this also is then 
furnished with a similar piece of tubing. By means of these pieces of tubing 
the mercurial thread can be placed in any desired position. 
According to the ordinary method, one extremity of the thread (say the 
left) is brought to the zero of the scale, and the position of its other extremity 
ascertained. By carefully shifting the thread the left extremity is brought 
exactly to the point where the right previously stood, and a new reading is 
made ; similar determinations are made all along the scale. This process is 
tedious in the extreme, on account of the great precision required at each 
placing of the thread. The following method,* which is equally accurate, 
* Since writing the above I have found evidence that this method was used by Bessex for ascer- 
taining positions (Pogg. Ann, 1xxxii. p. 290), 
VOL, XXIX. PART II, 7E 
